Zero-Shot Transfer
نقل بدون تدريب
قدرة النموذج على الأداء الجيد في مهام أو توزيعات بيانات لم يُدرَّب عليها صراحةً، معتمداً على المعرفة العامة المكتسبة أثناء التدريب المسبق.
The ability of a model to perform well on tasks or data distributions it was never explicitly trained on, relying on general knowledge learned during pre-training.
Also translated asالتعميم بدون أمثلة، التعميم بدون تدريب مُسبَق على المهمة، النقل الصفري
First appears in this corpus in: ALIGN: Scaling Up Visual and Vision-Language Representation Learning with Noisy Text Supervision (2021)
Appears in these papers
- ALIGN: Scaling Up Visual and Vision-Language Representation Learning with Noisy Text Supervision2021in the sky ✦
- BLIP-2: Bootstrapping Language-Image Pre-Training with Frozen Image Encoders and Large Language Models2023in the sky ✦
- Learning Transferable Visual Models from Natural Language Supervision2021in the sky ✦
- Learning Transferable Visual Models from Natural Language Supervision2021in the sky ✦
- Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection2023in the sky ✦
- ImageBind: One Embedding Space to Bind Them All2023in the sky ✦
- ImageBind: One Embedding Space to Bind Them All2023in the sky ✦
- Open X-Embodiment: Robotic Learning Datasets and RT-X Models2024in the sky ✦
- Segment Anything2023in the sky ✦
- Segment Anything2023in the sky ✦
- Self-Instruct: Aligning Language Models with Self-Generated Instructions2022in the sky ✦
- Sparks of Artificial General Intelligence: Early Experiments with GPT-42023in the sky ✦
- Voyager: An Open-Ended Embodied Agent with Large Language Models2023in the sky ✦